The project to be pursued concerns a new model of movement selection. The model is designed to explain how limb segments are recruited for reaching and other tasks. The assumed computations allow limb segments to compensate automatically for reduced mobility of other segments. In addition, the model explains established effects of practice, movement speed, and required accuracy, and predicts new effects to be tested experimentally. Although the planned research focuses on motor control, it draws upon and will contribute to other branches of cognitive and neural science, including those branches in which the evaluation of mental health relies heavily on the analysis of motor behavior.